| InternVideo |
- Top 1 Accuracy
- 70.0
|
InternVideo: General Video Foundation Models via Generative and Discriminative Learning
2022-12-06 |
코드 1 · 코드 2 |
| VideoMAE V2-g |
- Top 1 Accuracy
- 68.7
- Top 5 Accuracy
- 91.9
|
VideoMAE V2: Scaling Video Masked Autoencoders with Dual Masking
2023-03-29 |
코드 1 |
| Side4Video (EVA ViT-E/14 |
- Top 1 Accuracy
- 67.3
- Top 5 Accuracy
- 88.8
|
Side4Video: Spatial-Temporal Side Network for Memory-Efficient Image-to-Video Transfer Learning
2023-11-27 |
코드 1 · 코드 2 |
| ATM |
- Top 1 Accuracy
- 65.6
- Top 5 Accuracy
- 88.6
|
What Can Simple Arithmetic Operations Do for Temporal Modeling?
2023-07-18 |
코드 1 · 코드 2 |
| TAdaFormer-L/14 |
- Top 1 Accuracy
- 63.7
|
Temporally-Adaptive Models for Efficient Video Understanding
2023-08-10 |
코드 1 |
| TDS-CLIP-ViT-L/14(8frames) |
- Top 1 Accuracy
- 63.0
- Top 5 Accuracy
- 87.8
|
TDS-CLIP: Temporal Difference Side Network for Image-to-Video Transfer Learning
2024-08-20 |
코드 1 |
| UniFormerV2-L |
- Top 1 Accuracy
- 62.7
- Top 5 Accuracy
- 88.0
|
UniFormerV2: Spatiotemporal Learning by Arming Image ViTs with Video UniFormer
2022-09-22 |
코드 1 · 코드 2 |
| StructVit-B-4-1 |
- Top 1 Accuracy
- 61.3
|
Learning Correlation Structures for Vision Transformers
2024-04-05 |
연결 없음 |
| UniFormer-B (IN-1K + Kinetics400) |
- Top 1 Accuracy
- 60.9
- Top 5 Accuracy
- 87.3
- Param.
- 50.1
- GFLOPs
- 259x3
|
UniFormer: Unified Transformer for Efficient Spatial-Temporal Representation Learning
2021-09-29 |
코드 1 · 코드 2 · 코드 3 |
| TAdaConvNeXtV2-B |
- Top 1 Accuracy
- 60.7
|
Temporally-Adaptive Models for Efficient Video Understanding
2023-08-10 |
코드 1 |
| TPS |
- Top 1 Accuracy
- 58.3
|
Spatiotemporal Self-attention Modeling with Temporal Patch Shift for Action Recognition
2022-07-27 |
코드 1 |
| MSMA (8+16frames) |
- Top 1 Accuracy
- 57.9
|
Multi-scale Motion-Aware Module for Video Action Recognition
2023-02-19 |
연결 없음 |
| UniFormer-B (IN-1K + Kinetics600) |
- Top 1 Accuracy
- 57.6
- Top 5 Accuracy
- 84.9
- Param.
- 21.4
- GFLOPs
- 41.8x3
|
UniFormer: Unified Transformer for Efficient Spatial-Temporal Representation Learning
2021-09-29 |
코드 1 · 코드 2 · 코드 3 |
| SIFA |
- Top 1 Accuracy
- 57.3
|
Stand-Alone Inter-Frame Attention in Video Models
2022-06-14 |
코드 1 |
| EAN ResNet50 (single clip, center crop,8+16 ensemble, with sparse Transformer) |
- Top 1 Accuracy
- 57.2
- Top 5 Accuracy
- 83.9
|
EAN: Event Adaptive Network for Enhanced Action Recognition
2021-07-22 |
코드 1 |
| TCM (Ensemble) |
- Top 1 Accuracy
- 57.2
|
Motion-driven Visual Tempo Learning for Video-based Action Recognition
2022-02-24 |
코드 1 · 코드 2 |
| BQNEn (ImageNet + K400 pretrained) |
- Top 1 Accuracy
- 57.1
- Top 5 Accuracy
- 84.2
|
Busy-Quiet Video Disentangling for Video Classification
2021-03-29 |
코드 1 · 코드 2 |
| TDN ResNet101 (one clip, center crop, 8+16 ensemble, ImageNet pretrained, RGB only) |
- Top 1 Accuracy
- 56.8
- Top 5 Accuracy
- 84.1
|
TDN: Temporal Difference Networks for Efficient Action Recognition
2020-12-18 |
코드 1 |
| SELFYNet-TSM-R50En (8+16 frames, ImageNet pretrained, 2 clips) |
- Top 1 Accuracy
- 56.6
- Top 5 Accuracy
- 84.4
|
Learning Self-Similarity in Space and Time as Generalized Motion for Video Action Recognition
2021-02-14 |
코드 1 |
| CT-Net Ensemble (R50, 8+12+16+24) |
- Top 1 Accuracy
- 56.6
|
CT-Net: Channel Tensorization Network for Video Classification
2021-06-03 |
코드 1 |
| MoDS (8+16frames) |
- Top 1 Accuracy
- 56.6
|
Action Recognition With Motion Diversification and Dynamic Selection
2022-07-15 |
연결 없음 |
| MLP-3D |
- Top 1 Accuracy
- 56.5
|
MLP-3D: A MLP-like 3D Architecture with Grouped Time Mixing
2022-06-13 |
연결 없음 |
| RSANet-R50 (8+16 frames, ImageNet pretrained, 2 clips) |
- Top 1 Accuracy
- 56.1
- Top 5 Accuracy
- 82.8
|
Relational Self-Attention: What's Missing in Attention for Video Understanding
2021-11-02 |
코드 1 |
| SELFYNet-TSM-R50En (8+16 frames, ImageNet pretrained, a single clip) |
- Top 1 Accuracy
- 55.8
- Top 5 Accuracy
- 83.9
|
Learning Self-Similarity in Space and Time as Generalized Motion for Video Action Recognition
2021-02-14 |
코드 1 |
| RSANet-R50 (8+16 frames, ImageNet pretrained, a single clip) |
- Top 1 Accuracy
- 55.5
- Top 5 Accuracy
- 82.6
|
Relational Self-Attention: What's Missing in Attention for Video Understanding
2021-11-02 |
코드 1 |
| PAN ResNet101 (RGB only, no Flow) |
- Top 1 Accuracy
- 55.3
- Top 5 Accuracy
- 82.8
|
PAN: Towards Fast Action Recognition via Learning Persistence of Appearance
2020-08-08 |
코드 1 · 코드 2 |
| GSM Ensemble InceptionV3 (ImageNet pretrained) |
- Top 1 Accuracy
- 55.16
|
Gate-Shift Networks for Video Action Recognition
2019-12-01 |
코드 1 · 코드 2 |
| MSNet-R50En (ensemble) |
- Top 1 Accuracy
- 55.1
|
MotionSqueeze: Neural Motion Feature Learning for Video Understanding
2020-07-20 |
코드 1 · 코드 2 |
| AE-Net (8+16frames) |
- Top 1 Accuracy
- 55.0
|
AE-Net:Adjoint Enhancement Network for Efficient Action Recognition in Video Understanding
2022-07-21 |
연결 없음 |
| VoV3D-L (32frames, Kinetics pretrained, single) |
- Top 1 Accuracy
- 54.59
- Top 5 Accuracy
- 82.30
- Param.
- 5.8M
- GFLOPs
- 20.9x6
|
Diverse Temporal Aggregation and Depthwise Spatiotemporal Factorization for Efficient Video Classification
2020-12-01 |
코드 1 |